1. Field of the Invention
The present invention relates to a network address managing method and system for managing network (node) addresses at node positions in a network system which comprises ports connected to the network and nodes connectable to the ports in a plug-in manner.
2. Prior Art
FIG. 1 is a block diagram showing the configuration of a network system which adopts a conventional network address managing system. In FIG. 1, the reference numeral 11 denotes a plurality of ports connected to network cables N1, N2 for input/output of information, 12 a plurality of nodes connectable to the ports 11 in a plug-in manner, and 13 gate ways (or bridges) for interconnecting the network cables N1 and N2.
FIG. 2 is a representation showing a correspondence table 25 between node names and node addresses. The node 12 of a node name A is located at a node address 1-1, the node 12 of a node name B at a node address 1-2, the node 12 of a node name C at a node address 2-1, and the node 12 of a node name D at a node address 2-2, respectively.
Processing by means of that conventional network address managing system will now be described. When the nodes 12 are connected to the network cables N1, N2 via the ports 11, a management operator of the network assigns respective nodes their node addresses corresponding to node names, as exemplarily shown in the correspondence table 25 of FIG. 2, for uniquely identifying the nodes throughout the network. Sharing of the correspondence table 28 allows each node 12 to know the node address of the node 12 to be communicated and carry out communication with that node 12.
The above conventional network address managing system is problem free in the case that the network configuration is fixed and does not change. However, when the nodes are optionally connectable to the ports in a plug-in manner, the nodes can be easily moved from network to network, which leads to the problem described below. Namely, whenever any of the nodes is moved among networks, the management operator of the system must update the correspondence table between node names and node addresses for each network without causing a contradiction in the table and, as a result, the burden imposed on the management operator is increased.
More specifically, in FIG. 1, each node 12 incorporates therein a node address storage section 14 to store its own address, and is connected to the network cable N1 or N2 via the port 11. By connecting a number of nodes to the respective network cables in such a manner, the nodes can be distributed over the network.
Looking at a network environment including the nodes distributed therein, each node 12 holds in the node address storage section 14 a node address used for identifying itself in the network, and is also connected to the network cable N1 or N2 via the port 11.
Thus, the nodes 12 on the network are connected to the network cables N1, N2 via the respective ports 11 and hold in their node address storage sections 14 respective node addresses uniquely assigned throughout the network. While each node is identified on the network using its node address, the node is connected to the port in a plug-in manner and thus portable. When the node is connected to another network, the content of the node address storage section 14 is corrected because of a different node address assigned thereto.
Since the conventional ports just connect the nodes and the network cables, none of the nodes can identify to which port out of the numeral ones it is connected. In order to identify the node, therefore, it is required for the node to directly hold the node address corresponding to the port to which it is connected. Furthermore, in the case that the node is connected to another port in a plug-in manner and moved to another network, the node address must be updated so as to correspond to the port to which it has been newly connected.
To solve the above problem, there has so far been disclosed "an address determining system" in Japanese Patent Public Disclosure (Kokai) No. 1-166640. With this system, a star connection LAN system includes a network extension unit interconnecting respective terminals wherein a local address transferred to the terminal is initially set by setting a dip switch or the like. When a first terminal is connected to the LAN system, the initially set local address is transferred to the first terminal and when a next terminal is connected to the LAN system, the initially set local address is counted up and the resulting local address is transferred to the newly connected next terminal. As a result, there are a plurality of network extension units each of which is required to initially set the local address, and set a new local address by hardware means in need of updating to another local address. The incidental work or processing and address management become more intricate as the number of units increases.